Predictors of disease free survival in node negative ampullary adenocarcinoma

Abstract

Background: Pancreatico-duodenectomy is the only curative treatment option for patients with resectable ampullary adenocarcinoma (AA). Excellent disease free survival (DFS) can be achieved in patients with clear resection margins and no lymph node involvement. The potential role of adjuvant chemotherapy in this patient cohort is not fully understood. The aim of this study was to assess predictors of DFS in patients with node negative AA. Material & Methods: Consecutive patients undergoing pancreatico-duodenectomy for AA at our institution from 1996-2017 were included. Following institutional approval, demographic, surgical and histopathological data were analysed retrospectively. Independent pathological review confirmed diagnosis of AA. Patients were divided into lymph node positive and lymph node negative groups. Lymph node involvement was assessed by lymph node ratio (LNR). Using visual binning a threshold of 0.25 was selected for categorisation. Cox proportional hazard modelling was used to identify independent predictors of DFS. Factors with p<0.1 on univariate analysis were included in multivariate analysis. Results: 39 out of 108 included patients did not have nodal involvement (N0) (Table 1). Average DFS in the N0 group was 122±20 vs. 33±5 months in the N1 group (p<0.0001). Tumour stage T1/T2 (Hazard ratio[HR] 0.14, p=0.003) and lymph node count>12 (HR 3.94, p=0.047) were statistically significantly associated with improved DFS in the N0 group. In the N1 group, male gender (HR 0.4, p=0.003), absence of post-operative complications (HR 0.48, p=0.023), posterior margin R0 status (HR 0.45, p=0.046) and extent of nodal involvement<0.25 (HR 0.53, p=0.03) were associated with improved DFS. Conclusion: The majority of patients with lymph node negative AA has excellent long-term DFS. Advanced tumour stage and specimen lymph node number<12 were found to be independent predictors of DFS. In lymph node negative AA these predictors may allow identification of a patient subgroup that could benefit from adjuvant chemotherapy.Tabled 1